- From: Jonas Liljegren <jonas@paranormal.o.se>
- Date: Sun, 02 Jan 2000 21:50:56 +0100
- To: RDF Intrest Group <www-rdf-interest@w3.org>
- CC: Pierre-Antoine CHAMPIN <champin@cpe.fr>
Pierre-Antoine CHAMPIN wrote: > > Jonas Liljegren wrote: > > > > Pierre-Antoine CHAMPIN wrote: > > > And neither property-value-unicity nor node-unification are adressed by RDF recommandation. > > > That's a really interesting point, but still a big issue, IMHO > > > But we could prepare for this > > by handling unknown URIs diffrent from known URIs. > > Here again I don't quite agree : > Sergey pointed out that there is no fundamental difference between > anonymous and ID'ed resources, which I agree with. I wasn't talking about anonymous URIs, but of unknown URIs. A query could be a couple of statements with one or more of the URIs marked as the one, for wiitch we would like to know the URI aliases for. I suggested that this marking could be in the URI itself, by using a URI explicitly intended to be temporary. This would be analogous with the repeated use of the variables x, y and others in mathematical equations. But I could agree with that we maby better use type properties to mark the variables as unknown URIs. -- / Jonas - http://paranormal.o.se/perl/proj/rdf/schema_editor/
Received on Sunday, 2 January 2000 15:51:13 UTC